現今大部份程式開發皆已使用物件導向分析(Object-Oriented Analysis)和物件導向設計(Object-Oriented Design),以期能利用物件之繼承、封裝、多形等觀念,來縮短程式開發之時間及增加程式碼之再利用性。而雖然從系統分析到程式開發,皆以物件導向為主軸,再配合統一模式語言(UML)和設計樣版(Design Pattern)等相關技術構成一套相當完整的物件導向系統開發流程架構,但儲存資料之後端資料庫至今卻仍然大多數都使用關聯式資料庫(Relational Database)(物件資料庫雖然已有產品發表,但至今仍然不成熟),而變成了從系統分析至程式碼撰寫,皆利用物件導向技術,但存取後端之關連式資料庫時,卻必須使用關連式資料庫之表格(Table)、欄位(Column)等儲存單位及結構式查詢語言(Structural Query Language),而使得前端之物件和後端資料庫結構無法直接對應。 因此許多人提出了在關連式資料庫上面,以物件類別來封裝關連式資料庫之資料結構以及存取資料庫之動作,讓程式設計師可以以操縱物件的方式來存取關連式資料庫,而不必再去管後端資料庫如何存取資料,也不必在程式碼中使用結構式查詢語言,降低程式與資料庫的相依性。但截至目前為止,此種實作方式會限制程式語言的使用,如以Java封裝之資料庫物件,無法被Visual Basic使用,而只能限制程式設計師以Java來存取,更遑論以不同平台裝置來存取此一資料庫物件的可能性(如PDA、Tablet PC等…),而如果分公司之間的連線需透過防火牆,此一問題也會更形困難。 有鑑於此,本篇論文提出了一個解決的方法與架構,讓程式設計師可以以任何一種程式語言(支援Web Services的語言)來存取資料庫物件,而不同平台的裝置、甚至透過Internet穿越防火牆,也都能很容易的存取到此一資料庫物件。 Today most programs are developed using by OOA(Object-Oriented Analysis) and OOD(Object-Oriented Design) to reduce the time of developing systems and increase the code reuse. Although it is a complete development architecture for using OO technology with UML(Unified Modeling Language) and Design Patterns, but the database for storing data is still using RDB(Relational Database). Relational database uses tables and columns to store data and programmers must use the SQL(Structural Query Language) to access data. It is difficult to mapping objects to relational database data structure. For solving this problem, many people use a set of classes to encapsulate the relational database data structure. Programmers manipulate data through this layer of classes. Just like using object, programmers don’t need to know the data structure of relational database and don’t need to use the SQL to access data. But this solution restricts the using of programming languages. For example, if the classes are written by java, then programmers must use java to access this classes. In this thesis we proposed an architecture to solve the problems. With this architecture, programmers can use any programming languages which support Web Services to access the database objects that encapsulate the database data structures. It is also easy for programmers to access the database objects through firewall over Internet.